In this cross-sectional postal study, the authors measured nitrogen dioxide levels inside infants' bedrooms and outside their homes. During the 2-wk monitoring period, the authors investigated the association between nitrogen dioxide levels and 20 infant symptoms. The subjects were 1,200 women who had infants aged 3-12 mo. Median levels of indoor and outdoor nitrogen dioxide were 6.8 and 12.6 ppb, respectively. Environmental factors that were associated significantly with indoor levels were gas cooking, cigarette smoking, reported traffic levels, and presence of a kerosene heater; use of a cooker hood was associated negatively with indoor nitrogen dioxide levels. There was no evidence for any short-term significant association between prevalence of respiratory symptoms and nitrogen dioxide levels. Diarrhea, the only symptom associated significantly and positively with indoor nitrogen dioxide levels, had unadjusted and adjusted odds ratios of 1.48 (95% confidence interval: 1.13, 1.95) and 1.38 (1.11, 1.70), respectively. This association is discussed in terms of a proposed mechanism with nitric oxide. No association between a gas cooker in the home and diarrhea was found. The association between diarrhea and nitrogen dioxide level might have been a chance finding; the authors investigated 20 symptoms, and at least 1 was expected to be significant at the .05 level. The finding, however, was similar to that reported in a previous study in which a gas cooker was a proxy for nitrogen dioxide exposure.